The Influence of Coloring Liquids on the Substrate Masking Ability and Adhesive Properties of Different Zirconia Ceramics

ABSTRACT

Objective

The aim of this study was to assess the colorimetric characteristics and substrate masking ability of zirconia specimens with coloring liquids applied at the intaglio surface; the zirconia bond strength to a resin‐based luting agent and topographic characteristics were also assessed.

Materials and Methods

Zirconia materials tested were: 5Y‐PSZ (Prettau Anterior), 4Y‐PSZ (IPS e.max ZirCAD MT), and 3Y‐TZP (IPS e.max ZirCAD MO) veneered with lithium disilicate (IPS e.max CAD LT). Disc‐shaped specimens (8 mm of diameter, shade A2, n = 10) with 1.5 and 2.0 mm of thicknesses were produced for the colorimetric assessments. For the experimental groups, coloring liquids White and White Plus were applied at intaglio surface of zirconia specimens before sintering. Data for lightness (L*), chroma (C*), hue angle (h^o), translucency parameter (TP₀₀), and whiteness index for Dentistry (WI_D) were recorded; color differences (ΔE₀₀) of specimens over discolored substrates C3, C4, coppery and silvery metals in comparison with a reference (A2) were calculated. Additional specimens (n = 10) were produced to assess the microshear bond strength (μSBS) of a resin‐based luting agent to the zirconia intaglio surface with and without coloring liquid. The outcomes were obtained after 24 h water storage and after 10,000 thermal cycles (5°C–55°C). Analyses of variance (two‐way for L*, C*, and h^o and three‐way for TP₀₀, WI_D, ∆E₀₀, and μSBS) and Tukey post hoc test (α = 0.05) were applied. Topographic images were also obtained via scanning electron microscopy (SEM).

Results

The use of coloring liquids (especially White Plus) and thicker zirconia specimens provided, in general, reductions in TP_00, ΔE₀₀ and hue angle, and an increase in lightness and chroma (p < 0.001). WI_D increased in 5Y‐PSZ and decreased in 4Y‐PSZ with the use of coloring liquids in specimens of 1.5 mm (p < 0.001). The μSBS of groups was in the range from 20 to 25 MPa, in which only specimens of 5Y‐PSZ with coloring liquids showed statistically lower MPa values in comparison with the control group after aging (p > 0.05). In the SEM analysis, zirconia specimens with coloring liquids did not show relevant differences in topographic characteristics in comparison with controls.

Conclusion

The application of coloring liquids at zirconia intaglio surface has significantly improved its substrate masking ability without detrimental effects on the bond strength to a resin‐based luting agent and topographic characteristics.

Clinical Significance

The use of coloring liquids at zirconia intaglio surface before sintering is an effective technique to enhance the masking ability of restorations confectioned over discolored substrates.

1. Introduction

Zirconia has continuously emerged as a dental restorative material due to its chemical stability, biocompatibility, and adequate mechanical properties, especially fracture toughness [1, 2]. The introduction of digital workflows into restorative dentistry, initially with computer aided design/computer aided manufacturing (CAD/CAM) and, more recently, with tridimensional (3D) printing, has assisted in popularizing zirconia restorations [3].

The first versions of zirconia for dental restorations are basically composed by small grain‐sized zirconium dioxide (ZrO₂) particles stabilized mainly at tetragonal phase after sintering with 3 mol% of yttrium (3Y‐TZP) [2, 3]. Such material presents elevated strength, but also high opacity [4], and is traditionally used in dentistry as frameworks for multi‐unit restorations, initially veneered with porcelain and, more recently, with lithium disilicate in the CAD/CAM digital workflow, a technique known as CAD‐on [5]. The subsequent generations of zirconia have been produced with higher translucencies. Such aim has been achieved by increasing the grain size of ZrO₂, the amount of cubic phase into its structure, and the yttrium content to 4 and 5 mol% [2, 3]. Such modifications in the material composition have made possible the use of zirconia as monolithic esthetic restorations [3].

One common challenge in restorative dentistry is the presence of discolored abutment substrates. Previous studies have shown limited potential of restorative materials, including zirconia, to present an adequate balance of translucency (to mimic the natural dental structures) and opacity (to mask such discolored substrate) [6, 7, 8, 9, 10, 11, 12, 13]. Therefore, development of materials and techniques has been made in such direction.

A recent technique has been presented for improving the substrate masking ability of zirconia without negatively interfering with the final restoration aspect [14]. Such technique consists of the application of coloring liquids at the zirconia intaglio surface before the sintering process, which infiltrates the restoration structure and creates an opaque internal layer after sintered [14]. The quantification of the zirconia masking ability with this technique, based on properties such as translucency parameter and eventual alterations in the colorimetric characteristics of the restoration, becomes of interest. Moreover, before the clinical popularization of the technique, it is pertinent to evaluate the influence of the coloring liquid on the zirconia bonding ability to resin‐based luting agents. Zirconia is considered a low‐reactive and acid‐resistant material, characteristics that have, for years, complicated a clinically relevant bond strength to resin‐based luting agents [1, 2, 3]. Surface treatments have been continuously developed, resulting in recent techniques with clinically acceptable bond strength values [15]. Therefore, the use of new products at the zirconia intaglio surface should not overcome such improvements. Possible significant alterations in ZrO₂ grain characteristics, such as grain size, caused by the use of coloring liquids should also be evaluated [16]. Irregularities, such as grain detachments and the presence of surficial defects or porosity, should be investigated [16, 17]. These alterations in the material characteristics could affect either the material's bulk resistance or the bond strength by serving as origins for crack propagation and cohesive failures [16, 17, 18, 19].

Therefore, the present study aimed to assess and quantify the effect of using coloring liquids at zirconia intaglio surface on colorimetric characteristics of zirconia specimens and the masking ability of discolored substrates. Moreover, effects of coloring liquid application on the zirconia bond strength to a resin‐based luting agent and topography (grain characteristics) were also assessed. The tested hypotheses of the study were that the use of coloring liquids at zirconia intaglio surface could influence (I) the colorimetric properties of lightness, chroma, hue angle, and whiteness index for dentistry, (II) the translucency parameter and the substrate masking ability (based on data of color differences), (III) the bond strength to a resin‐based luting agent, and (IV) the zirconia topographic characteristics.

2. Materials and Methods

2.1. Experimental Design and Materials Information

The study flowchart and the representative images of the specimens are described in Figures 1 and 2, respectively. The description of ceramic materials, commercial name, manufacturer, and composition are presented in Table 1. The 3Y‐TZP is a second‐generation zirconia, while 4Y‐ and 5Y‐PSZ are materials of the third generation.

FIGURE 1.

Experimental design of the study.

FIGURE 2.

(A) Zirconia specimens before sintering; (right) experimental specimen, the red pigment was mixed with the coloring liquid by the manufacturer for serving as parameter of amount and distribution of the product by the operator; (left) control specimen. (B) Images of intaglio surface of sintered zirconia specimens with coloring liquids and their control. (C) Representative images of masking ability of zirconia specimens (4Y‐PSZ, 1.5 mm of thickness).

TABLE 1.

Description of ceramic materials, commercial name, manufacturers, and composition.

Material	Commercial name, manufacturer, and batch number	Composition (wt%)
5Y‐PSZ	Prettau Anterior, Zirkonzahn SRL, Gais, Italy. ZB6111D	ZrO2 (main component), Y2O3 < 12%, Al2O3 < 1%; SiO2 max 0.02%; FeO3 max 0.02%
4Y‐PSZ	IPS e.max ZirCAD, MT A2, Ivoclar AG, Schaan, Liechtenstein. Z00FLZ	ZrO2 86.0%–93.5%; Y2O3 6.5%–8.0%; HfO2 ≤ 5.0%; Al2O3 ≤ 1.0%; other oxides ≤ 1.0%.
3Y‐TZP (coping)	IPS e.max ZirCAD, MO, Ivoclar AG, Schaan, Liechtenstein. T42432	ZrO2 88.0%–95.5%; Y2O3 4.5%–6.0%; HfO2 ≤ 5.0%; Al2O3 ≤ 1.0%; other oxides ≤ 1.0%.
Lithium disilicate ceramic (veneer)	IPS e.max CAD, LT, A2, Ivoclar AG, Schaan, Liechtenstein. V13882	SiO2, Li2O, K2O, P2O5, ZrO5, ZnO, Al2O3, MgO, pigments
Coloring liquids for zirconia	White and White Plus, DMAX Co. Ltd, Daegu, South Korea. DMS210777WH DMS190612WP	Water, nickel chloride hexahydrate, iron nitrate nonahydrate, sodium silicate, manganese nitrate hexahydrate, copper nitrate trihydrate, chromium nitrate nonahydrate

2.2. Colorimetric Properties and Translucency Parameter

The sample size was calculated with the data of color differences (primary outcome) from the pilot study, in a statistical software (G*power v. 3.1.9.7; Heinrich‐Heine‐Universität Düsseldorf, Germany). The defined parameters were: effect size f = 0.9, α = 0.05, power = 0.80, actual power = 0.95, resulting in a minimum number of four specimens per group. To ensure statistical power for the other 05 outcomes obtained from the same specimens (L*, C*, h^o, TP₀₀, WI_D), 10 specimens were produced for each group (n = 10).

For measurement of colorimetric properties and translucency parameter, presintered zirconia blocs were shaped into cylinders of 10 mm diameter (to account for 20% contraction during the sintering process, resulting in specimens with 8 mm diameter), with a diamond disc (Coarse Grinding Discs, green—240 μm grit; Buehler, Like Bluff, USA) in a polishing machine (EcoMet/AutoMet 250; Buehler, Like Bluff, USA). Specimens were cut from the cylinders with a diamond blade in a cutting machine (Isomet 1000; Buehler, Like Bluff, USA). The surface topography was standardized with silicon carbide (SiC) papers (#400‐grit, #600‐grit, #1200‐grit, and #2500‐grit), and the specimen thickness measured with a digital caliper. The specimens were numerated and aleatorily distributed along the groups using random.org.

The application of coloring liquids (White and White Plus) at zirconia intaglio surface was carefully performed with a number 2 brush, with cautious control over the quantity and distribution over the specimen. The procedures were performed by a trained dental ceramist. All zirconia specimens were sintered with the standard sintering process of 9 h, with a maximum temperature of 1550°C in a holding time of 2 h, in an adequate furnace (Vita Zyrcomat T, Vita Zahnfabrik, Bad Säckingen, Germany). The specimens of 3Y‐TZP + veneer were confectioned by joining lithium disilicate veneers (0.5 mm) in a pre‐crystalline phase to a sintered zirconia copping (1 and 1.5 mm) with a fusion glass–ceramic (IPS e.max CAD Crystall./Connect, Ivoclar AG, Schaan, Liechtenstein) at 850°C (Programat EP 5000‐G2, Ivoclar AG, Schaan, Liechtenstein).

For the confection of substrates (8 mm of diameter and 5 mm of thickness), resin composites (Filtek Z350^XT Dentin; 3 M ESPE, St Paul, USA) of shades A2, C3, and C4 were applied into a silicone mold covered by two glass slices, and each side was light‐activated with 1200 mW/cm² (Radii‐Cal; SDI, Victoria, Australia) for 40 s. Metallic substrates were produced by the casting method using acrylic resin patterns. Coppery substrates were obtained with copper‐aluminum alloy (Duracast MS Soft; DuraWear Corp, Birmingham, USA), and silvery substrates were obtained in cobalt‐chromium alloy (Duceralloy; Degudent gmbh/Dentsply, Hanau, Germany).

Color coordinates (L*, a*, and b*) of zirconia specimens were obtained with a spectrophotometer (SP60, EX‐Rite; Grand Rapids, USA) using a gray background (Flexipalette Color Match, Smile line, St‐Imier, Switzerland) in all measurements. The measurements were made under the following parameters: aperture setting 8 mm; illumination 12 mm; and measuring time of 2 s. The wavelength range was of l = 400–700 nm/intervals of 10 nm. Commission Internationale de l'Eclairage (CIE)Lab color coordinates were calculated as per the CIE D65 Standard Illuminant and CIE 2‐degree standard observer [20, 21].

Chroma (C*) and hue angle (h^o) were calculated for each specimen [22]. Color differences (∆E₀₀) were obtained through the CIEDE2000 formula [22, 23] with data of zirconia specimens tested over discolored substrates (C3, C4, coppery and silvery metals) in comparison with the data obtained over the referential substrate (A2); a try‐in paste (Allcem Veneer, shade A2; FGM Dental Group, Joinville, Brazil) was used as coupling agent [24]. After each measurement, the specimens were cleaned with 78% isopropyl alcohol. The CIEDE2000 formula was applied, as follows: [22, 23]

$${\mathit{\Delta E}}_{00}={\left[{\left(\frac{∆L^{\prime }}{K_L{S}_L}\right)}^2+{\left(\frac{∆C^{\prime }}{K_C{S}_C}\right)}^2+{\left(\frac{∆H^{\prime }}{K_H{S}_H}\right)}^2+R_T\left(\frac{∆C^{\prime }}{K_C{S}_C}\right)\left(\frac{∆H^{\prime }}{K_H{S}_H}\right)\right]}^{\frac{1}{2}}$$

where ΔL′, ΔC′, and ΔH′ are the differences in luminosity (L′), chroma (C′), and hue (H′), respectively, for a pair of measurements. R _T is a rotation function that accounts for the interaction between chroma and hue differences in the blue region. S _L , S _C , and S _H are weighting functions that adjust the total color difference for variation in the location of the color difference pair in the L*, a*, b* coordinates, and the parametric factors k _L , k _C , and k _H are correction terms for deviation from reference experimental conditions [23]. The parametric factors were set as 1 [22].

The ∆E₀₀ values obtained from the zirconia specimens over the A2 substrate and the other shade substrates were compared with the color difference visual thresholds previously reported in literature: perceptibility threshold (PT) (∆E₀₀ = 0.81), acceptability threshold (AT) (∆E₀₀ = 1.77), and mismatch type A—moderately unacceptable (∆E₀₀ = 3.60) [25, 26].

Color coordinates (L*, C*, and h^o) [22] of substrates and the ∆E₀₀ [22, 23] between each of the discolored substrates and the referential (A2) were also measured (Table 2).

TABLE 2.

Mean ± standard deviation (±SD) of lightness (L*), chroma (C*), and hue angle (h°) of substrates, and color difference (∆E₀₀) in comparison with the A2 reference.

Substrate	L* ± SD	C* ± SD	h° ± SD	∆E00 / A2
A2	72.34 ± 0.00	11.58 ± 0.02	89.73 ± 0.07	—
C3	63.54 ± 0.01	6.90 ± 0.05	95.92 ± 0.18	6.94
C4	56.70 ± 0.00	10.11 ± 0.02	90.96 ± 0.07	14.66
Coppery	62.34 ± 0.04	16.79 ± 0.01	88.30 ± 0.03	12.73
Silvery	51.20 ± 0.01	0.42 ± 0.01	256.87 ± 0.94	25.84

The translucency parameter (TP₀₀) [27] of zirconia specimens was calculated measuring the differences of color coordinates over black (L* = 26.37, a* = 0.17, b* = 0.36) and white (L* = 90.94, a* = −0.94, b* = 6.00) backgrounds. Differences in TP₀₀ (∆TP₀₀) between specimens with coloring liquids and controls were compared with perceptibility (TPT = 0.62) and acceptability (TAT = 2.62) thresholds [28]. In order to avoid light dispersion between the zirconia specimen and the background, caused by differences in refractive index, glycerol was used as a coupling solution.

Whiteness index for Dentistry (WI_D) of zirconia specimens was calculated with color coordinates obtained over the black background [29]. Differences in WI_D (∆WI_D) between specimens with coloring liquids and controls were compared with perceptibility (WPT = 0.72) and acceptability (WAT = 2.60) thresholds [30]. WI_D formula was applied as follows: [29]

$${\mathit{WI}}_D=0.511L^{*}-2.324a^{*}-1.100b^{*}$$

2.3. Microshear Bond Strength (μSBS)

For μSBS, sample size was calculated with the data from the pilot study in a statistical software (G*power v. 3.1.9.7; Heinrich‐Heine‐Universität Düsseldorf, Germany). The defined parameters were: effect size f = 0.71, α = 0.05, power = 0.80, actual power = 0.87, resulting in a minimum number of eight specimens per group; 10 zirconia specimens were produced to ensure adequate significance. Forty rectangular specimens (12 × 12 × 5 mm) were produced for each zirconia and numerated; twenty of them were randomly distributed to the experimental group (random.org), which received the coloring liquid White Plus, and the remaining specimens served as controls. After the sintering process (as mentioned previously), all zirconia specimens were placed into polyvinyl chloride plastic rings and embedded with polymethyl methacrylate (JET; Clássico, São Paulo, Brazil). The zirconia bonding surface was sandblasted with aluminum oxide particles (45 μm) at a distance of 10 mm with 2.8 bar pressure and cleaned in an ultrasonic bath with 70% isopropyl alcohol for 5 min [15]. All specimens received the active application of a coupling agent (Monobond N; Ivoclar AG, Schaan, Liechtenstein) for 15 s, left to react for another 45 s [15].

For the confection of resin‐based luting agent cylinders, starch matrices (n = 80, four for each zirconia specimen; height = 1 mm; internal diameter = 1.13 mm) (Isabela; M. Dias Branco S.A. Indústria e Comércio de Alimentos, Bento Gonçalves, Brazil) were fixed with sticky wax (Lysanda, São Paulo, Brazil) over the treated ceramic surfaces [31, 32]. A dual‐polymerizing resin‐based luting agent (Multilink N; Ivoclar AG, Schaan, Liechtenstein) was mixed and inserted into the matrices with the aid of a dental explorer and light activated for 40 s at 1200 mW/cm² (Radii‐Cal; SDI, Victoria, Australia). After 24 h in water storage (37°C), the matrices were carefully removed, and the luting agent cylinder was examined for the eventual presence of defects. In case of any irregularity, the specimen was replaced. Ten zirconia specimens were tested after storage in distilled water at 37°C for 24 h (baseline) and the other 10 were subjected to aging process at 10,000 thermocycles in water bath (5°C–55°C) [33].

For μSBS test, the wire‐loop method was used. A thin stainless‐steel wire (Ø = 0.2 mm) was looped around each luting agent cylinder at the bonding interface and tested in a universal testing machine (EMIC DL‐2000; EMIC, São José dos Campos, Brazil) with a crosshead speed of 1 mm/min until failure. μSBS (MPa) was calculated by dividing the load at failure (N) by the bonded interface area (1 mm²). After the test, all specimens were analyzed in a stereomicroscope (Stereo Discovery V20; Carl Zeiss, Oberkochen, Germany) at ×50 magnification to classify the failures into adhesive (failure evolving ≥ 50% of the adhesive interface) or cohesive (failure evolving ≥ 50% within the luting agent).

2.4. Topographic Analysis

Two additional rectangular specimens (10 × 10 × 3 mm) of each type of zirconia were produced, as mentioned earlier, for the analysis of topography (grain characteristics). One specimen was treated with the coloring liquid White Plus and the other served as control. The specimens were sputter‐coated with gold and analyzed at ×10,000 magnification in a scanning electron microscope (SEM; VEGA‐3G; Tescan, Brun, Czech Republic).

2.5. Statistical Analysis

Statistical analyses were performed by a blinded operator in a specific software (IBM SPSS Statistics, v.21; IBM Corp, New York, USA). Normality and homogeneity of data were confirmed with Shapiro–Wilk and Levene tests, respectively (p > 0.05). Two‐way ANOVA and Tukey post hoc test (α = 0.05) were used to detect significances in L*, C*, and h^o for each type of zirconia, considering the factors “coloring liquid” and “specimen thickness.” Three‐way ANOVA and Tukey post hoc test (α = 0.05) were used to compare the outcomes of TP₀₀, WI_D, ∆E₀₀ (factors: “type of zirconia,” “specimen thickness,” and “coloring liquid”) and zirconia bond strength to a resin‐based luting agent (factors: “type of zirconia,” “coloring liquid,” and “aging”).

3. Results

3.1. Colorimetric Properties and Translucency Parameter

The use of coloring liquids at zirconia intaglio surface and thicker specimens led, in general, to a significantly increased lightness and chroma, and a tendency for a reduction of hue angle, for all types of zirconia (p < 0.001) (Table 3).

TABLE 3.

Mean and standard deviation (±SD) of lightness (L*), chroma (C*), and hue angle (h°) of ceramic specimens.

Restorative material	Coloring liquid	Ceramic thickness (mm)	L* ± SD	C* ± SD	h° ± SD
5Y‐PSZ	Control	1.5	73.91 ± 0.49C	9.59 ± 0.93B	99.59 ± 0.86A
	White		79.72 ± 1.32B	9.16 ± 3.23B	99.42 ± 1.33A
	White plus		84.28 ± 0.56A	12.27 ± 0.30AB	97.38 ± 0.76A
5Y‐PSZ	Control	2.0	80.40 ± 2.30B	13.55 ± 1.55A	94.32 ± 1.33B
	White		83.61 ± 1.56A	14.76 ± 0.71A	94.00 ± 0.71B
	White plus		85.61 ± 1.20A	16.16 ± 0.57A	93.17 ± 0.84B
4Y‐PSZ	Control	1.5	74.76 ± 0.69D	9.11 ± 0.69C	97.18 ± 0.98A
	White		77.85 ± 0.95C	11.75 ± 0.58B	93.38 ± 0.87B
	White plus		80.20 ± 0.82B	13.06 ± 1.12B	92.72 ± 1.31B
4Y‐PSZ	Control	2.0	79.40 ± 0.74B	12.13 ± 1.02B	91.54 ± 1.73 bc
	White		80.74 ± 0.38AB	12.84 ± 0.67B	91.25 ± 0.79 bc
	White plus		81.61 ± 0.45A	14.41 ± 0.76A	89.49 ± 0.69C
3Y‐TZP + veneer	Control	1.5	76.86 ± 1.31D	10.15 ± 0.56B	96.41 ± 0.41A
	White		80.76 ± 0.59C	9.97 ± 0.49B	95.58 ± 0.72AB
	White plus		83.03 ± 0.54B	11.12 ± 0.61AB	94.52 ± 0.49 bc
3Y‐TZP + veneer	Control	2.0	81.71 ± 0.77 bc	10.31 ± 0.39B	93.91 ± 0.92C
	White		82.79 ± 0.38B	10.82 ± 0.57AB	94.52 ± 0.93 bc
	White plus		84.73 ± 0.34A	11.97 ± 0.61A	94.04 ± 0.57C

Note: Different uppercase letters in each column indicate statistical difference within each type of zirconia, according to two‐way ANOVA and Tukey's post hoc test, considering the factors “specimen thickness” and “coloring liquid.”

The whiteness index for Dentistry (WI_D) (Table 4) did not show significant differences with the use of coloring liquids in specimens of 2 mm of thickness (except the lower value of 4Y‐PSZ associated with White Plus) (p > 0.05). However, for specimens of 1.5 mm of thickness, WI_D values were lower for 4Y‐PSZ and higher for 5Y‐TZP with the use of coloring liquids (p < 0.001). The veneered 3Y‐TPZ ceramic showed a general tendency of higher WI_D values among the three types of zirconia, especially with 2 mm of thickness. ∆WI_D values (Table 4) were below WAT and TPT for specimens of 1.5 mm and 2 mm of thickness, respectively (with exception for the group 4Y‐PSZ/2 mm/white plus that presented ∆WI_D below WAT).

TABLE 4.

Mean and standard deviation (±SD) of whiteness index for Dentistry (WI_D) of the ceramic specimens.

Coloring liquid	Ceramic thickness (mm)	5Y‐PSZ	4Y‐PSZ	3Y‐TZP + veneer
WID
Control	1.5	31.05 ± 0.76 bc	30.89 ± 0.78 bc	30.82 ± 0.67 bc
White		33.49 ± 0.74A	28.45 ± 1.06D	32.60 ± 0.77AB
White plus		33.25 ± 0.11A	28.31 ± 0.86DE	32.25 ± 0.59AB
Control	2.0	28.57 ± 1.78D	27.99 ± 1.52D	32.06 ± 0.73AB
White		28.92 ± 0.97CD	27.78 ± 1.07D	32.41 ± 0.76AB
White plus		28.25 ± 0.91D	25.54 ± 1.12E	31.98 ± 0.89AB
∆WID
Control	1.5	ref	ref	ref
White		2.44	2.44	1.78
White plus		2.20	2.58	1.43
Control	2.0	ref	ref	ref
White		0.35	0.21	0.35
White plus		0.32	2.45	0.08

Note: Differences in WI_D (∆WI_D) of specimens with coloring liquids in comparison with controls form the same thickness and material. Uppercase letters represent statistical differences according the factors: type of zirconia, specimen thickness, and coloring liquid, depicted by three‐way ANOVA and Tukey post hoc test. Perceptibility threshold, WPT = 0.72. Acceptability threshold, WAT = 2.60. ref. = referential. Bold data = below WPT. Underlined data = below WAT.

Translucency parameter (TP₀₀) (Table 5) had an overall decrease with the use of coloring liquids (p < 0.001), especially for the most translucent ceramics (5Y‐ and 4Y‐PSZ), thinner specimens (1.5 mm), and the use of coloring liquid White Plus. Both groups of 4Y‐PSZ and 3Y‐TZP + veneer with coloring liquid White presented ∆TP₀₀ values below TPT and TAT with specimens of 2 and 1.5 mm of thickness, respectively. The same ceramics with coloring liquid White Plus presented ∆TP₀₀ values below TAT in specimens of 2 mm of thickness. For 5Y‐PSZ, only the use of coloring liquid White in specimens of 2 mm of thickness led to ∆TP₀₀ below TAT.

TABLE 5.

Mean and standard deviation (±SD) of translucency parameter (TP₀₀) for the ceramic specimens.

Coloring liquid	Ceramic thickness (mm)	5Y‐PSZ	4Y‐PSZ	3Y‐TZP + veneer
TP00
Control	1.5	10.72 ± 0.58A	9.10 ± 0.53B	7.62 ± 0.87 bc
White		6.61 ± 0.94CDE	6.76 ± 0.72CD	5.15 ± 0.27EFG
White plus		3.74 ± 0.29GHI	5.19 ± 0.76EFG	3.87 ± 0.40GHI
Control	2.0	5.74 ± 1.54DEF	5.20 ± 0.89EFG	4.68 ± 0.40FGH
White		4.00 ± 0.79GHI	4.59 ± 0.35FGH	4.10 ± 0.24GHI
White plus		2.73 ± 0.53I	3.35 ± 0.46HI	2.95 ± 0.22I
∆TP00
Control	1.5	ref	ref	ref
White		4.11	2.34	2.47
White plus		6.98	3.91	3.75
Control	2.0	ref	ref	ref
White		1.74	0.61	0.58
White plus		3.01	1.85	1.73

Note: Differences in TP₀₀ (∆TP₀₀) caused by the use of coloring liquids in comparison with control for specimens of the same thickness and material. Uppercase letters represent statistical differences according the factors: type of zirconia, specimen thickness, and coloring liquid, depicted by 3‐way ANOVA and Tukey post hoc tests. Perceptibility threshold, TPT = 0.62. Acceptability threshold, TAT = 2.62. ref. = referential. Bold data = below TPT. Underlined data = below TAT.

The analysis of color differences (ΔE₀₀) of zirconia specimens over discolored substrates in comparison with the reference revealed significant influence of factors type of zirconia, specimen thickness, and coloring liquid (p < 0.001). For substrate C3 (Figure 3A), specimens of 1.5 mm of thickness with coloring liquids showed ΔE₀₀ below AT. Ceramic specimens of 2 mm of thickness with coloring liquid White Plus (specially 5Y‐PSZ) showed ΔE₀₀ closer to that of PT. For substrate C4 (Figure 3B), ΔE₀₀ below AT was obtained with ceramic specimens of 2 mm of thickness associated with coloring liquid White Plus. Selected groups of specimens of 1.5 mm of thickness associated with White Plus (5Y‐PSZ and veneered 3Y‐TZP) also showed ΔE₀₀ close to AT.

FIGURE 3.

Mean and standard deviation of color differences (ΔE₀₀) for zirconia specimens tested over discolored substrates (C3, C4, coppery and silvery metals) in comparison with A2 (reference). Different letters indicate significance according to three‐way ANOVA and Tukey's post hoc test (α = 0.05) for the factors “type of zirconia,” “specimen thickness,” and “coloring liquid.” Dotted orange line, PT (0.81); red line, AT (1.77); blue line, mismatch type A (3.60).

For the substrate coppery metal (Figure 3C), ceramic specimens of 2 mm of thickness associated with coloring liquid White Plus showed ΔE₀₀ values close to AT (for 5‐YPSZ) and below AT (for 3Y‐TZP + veneer). Specimens of 1.5 mm of thickness with both coloring liquids showed ΔE₀₀ within the mismatch type A. For the substrate silvery metal (Figure 3D), ΔE₀₀ below the mismatch type A was obtained with the use of coloring liquid White Plus associated with zirconia specimens of 1.5 mm (5Y‐PSZ and 3Y‐TZP + veneer) and 2 mm of thickness.

3.2. Microshear Bond‐Strength

The μSBS (Table 6) showed a significant influence of factors type of zirconia, coloring liquid, and aging process (p < 0.001). For the baseline analysis, the zirconia specimens that received coloring liquids did not show significant differences in comparison with their respective controls (p > 0.05). After aging, only specimens of 5Y‐PSZ with coloring liquids showed a statistical reduction in comparison either with its control group or with the baseline (up to 20% for both) (p < 0.001). However, the microshear bond strength of all groups was in the range from 20 to 25 MPa. Predominantly adhesive failures (≥ 87%) were observed in all groups (Table 7).

TABLE 6.

Mean, standard deviation (SD), and 95% confidence interval (95 CI) of shear bond strength (MPa) at baseline and after aging.

Groups		Baseline		Aged		Decrease (%)
		Mean (SD)	95 CI	Mean (SD)	95 CI
5Y‐PSZ	Control	25.01 (4.72)A	23.46–26.56	24.58 (4.80)A	22.98–26.18	1.8
	Coloring liquid	25.26 (4.91)A*	23.65–26.87	20.35 (5.80)B*	18.42–22.29	19.4
4Y‐PSZ	Control	23.83 (5.34)AB	22.08–25.59	20.36 (5.78)AB	18.28–22.45	14.6
	Coloring liquid	19.87 (5.29)B	18.08–21.66	20.22 (5.45)B	18.43–22.01	−1.8
3Y‐TZP + veneer	Control	21.38 (6.12)AB	19.30–23.45	21.10 (5.40)AB	19.30–22.91	1.3
	Coloring liquid	20.65 (4.90)B	19.01–22.28	22.65 (7.08)AB	20.21–25.08	−9.7

Note: Different uppercase letters indicate statistical differences for the factors “type of zirconia” and “coloring liquid” treatment, according to three‐way ANOVA within the same storage condition (α = 0.05). Symbols (*) indicate statistical differences between baseline and aged groups, according to three‐way ANOVA (α = 0.05).

TABLE 7.

Descriptive analysis of failure mode.

	Groups	Aging condition	A	C	PTF
5Y‐PSZ	Control	Baseline	38 (95%)	2 (5%)	0 (0%)
		Aged	37 (93%)	3 (7%)	0 (0%)
	Coloring liquid	Baseline	38 (95%)	2 (5%)	0 (0%)
		Aged	37 (93%)	3 (7%)	0 (0%)
4Y‐PSZ	Control	Baseline	38 (95%)	2 (5%)	0 (0%)
		Aged	32 (80%)	3 (7%)	5 (13%)
	Coloring liquid	Baseline	36 (90%)	4 (10%)	0 (0%)
		Aged	38 (95%)	2 (5%)	0 (0%)
3Y‐TZP + veneer	Control	Baseline	36 (90%)	4 (10%)	0 (0%)
		Aged	37 (93%)	2 (5%)	1 (2%)
	Coloring liquid	Baseline	37 (93%)	3 (8%)	0 (0%)
		Aged	35 (87%)	5 (13%)	0 (0%)

Note: A—predominantly adhesive; C—predominantly cohesive within the resin cylinder; PTF—pre‐test failures.

3.3. Topography

The topographic features and grain characteristics of zirconia specimens treated with coloring liquids did not show relevant differences in comparison with controls in the qualitative SEM analysis (Figure 4).

FIGURE 4.

Scanning electron microscopy images of topographic characteristics of zirconia specimens confectioned with and without coloring liquids.

4. Discussion

The first hypothesis of this study, which the use of coloring liquids could influence the lightness, chroma, hue angle, and whiteness index for dentistry, was accepted. Increases of lightness and chroma and reduction of hue angle were observed with the association of zirconia and coloring liquids (p < 0.001). Influences on color characteristics can be explained by the whitish and opaque nature of coloring liquids; despite their application being restricted to the intaglio surface, the light reflection from such layer through the specimen can influence the shade of restoration [32, 34]. Such outcomes are important warnings to dental ceramists for the potential detrimental effect of coloring liquids on the final color of restorations if they are applied in excesses.

The influence of coloring liquids on WI_D was dependent on the specimen thickness (p < 0.001). Specimens of 2 mm thickness have not been relevantly influenced by the application of coloring liquids, and that might be anticipated by the light scattering into such thick bulk structure before providing relevant WI_D alteration [32]. For thinner specimens (1.5 mm), light was reflected from the internal opaque layer through the ceramic structure and affected WI_D [32]. However, ∆WI_D values of specimens were below the clinical thresholds reported in literature [30], even in specimens of 1.5 mm thickness, which means that the proposed technique would not be clinically detrimental to WI_D in the present conditions.

The second hypothesis of this study, which the use of coloring liquids at intaglio surface could influence the substrate masking ability of zirconia, was accepted. Zirconia specimens containing coloring liquids, especially White Plus, showed lower ΔE₀₀ over all discolored substrates in comparison with the controls. This fact might be explained by the lower TP₀₀ of zirconia specimens containing coloring liquids (Table 5) [32]. The association of TP₀₀ and ΔE₀₀ is well reported in literature, with strong positive correlation between the two factors [6, 10, 12]. The use of thicker zirconia specimens showed significant reduction of ΔE₀₀ (p < 0.001). Such outcomes are also attributed to the lower TP₀₀ of the respective specimens (Table 5) [6, 20, 32]. This positive association of TP₀₀ and specimen thickness is also in agreement with previous studies [6, 20, 32]. The ceramic type has shown, in general, significant influence on TP₀₀ and ΔE₀₀ in specimens without coloring liquids, for which significantly higher values were observed for 5Y‐PSZ and lower values for 3Y‐TZP + veneer (p < 0.001). This was initially expected due to the increased cubic phase and larger grain‐sizes of 5Y‐ and 4Y‐PSZ in comparison with 3Y‐TZP [2, 3, 4]. However, these differences in zirconia composition have been surpassed by the use of coloring liquids and the formation of the opaque layer at intaglio surface, which led, in general, to an overall reduction in TP₀₀ and provided similar outcomes in the experimental groups among the different types of zirconia (Table 5).

Considering data of ∆TP₀₀, influences from the ceramic type and thickness, and from the type of coloring liquid were observed. 5Y‐PSZ with coloring liquids had the greatest ∆TP₀₀, and this can be explained by the presence of larger grain sizes and higher cubic phase in its structure [1, 2, 3], which allows greater influence of light transmission and reflection from the opaque internal layer. Specimens of 1.5 mm of thickness also provided higher ∆TP₀₀ than that of 2 mm due to their thinner bulk structure and, consequently, lower scattering of the light reflected from the substrate [32]. Therefore, the use of coloring liquids should be made with caution because the relevant number of ∆TP₀₀ values above TAT [28]. The coloring liquid White provided lower ∆TP₀₀ values than White Plus, with general outcomes below TAT [28], and its use should be considered in cases which the ∆TP₀₀ is of significant clinical relevance.

It is worth mentioning that color matching, using as references AT and PT thresholds, was dependent on the severity of substrate discoloration (Figure 3). Metallic substrates, especially the silvery‐shaded, caused the overall higher color differences (p < 0.001). This influence of substrate discoloration on the masking ability can be explained by the results presented in Table 2, for which silvery metals showed the highest ΔE₀₀ in comparison with the reference (A2), and substrate C3 showed the lowest ΔE₀₀ [11]. To achieve color matching, the interdependence of substrate shade and the characteristics of restorative material (such as TP₀₀) was also observed previously [11]. Therefore, in the clinical scenario, the shade of substrate must be recorded firstly, to guide the subsequent decisions of material selection and restoration thickness.

The third hypothesis of this study was accepted, as the use of coloring liquids at zirconia intaglio surface influenced the microshear bond strength (μSBS) to a resin‐based luting agent (p < 0.001); however, it is worth mentioning that only 5Y‐PSZ had a statistically significant reduction in μSBS with the use of coloring liquids, and that all groups showed μSBS in the range from 20 to 25 MPa, denoting no relevant influences of coloring liquid on the μSBS, as seen in Table 6. This might, in part, be explained by the SEM images (Figure 4) comparing surfaces of zirconia specimens with and without coloring liquids, in which no differences could be depicted in surface features and grain characteristics between groups [16, 17, 18, 19]. Moreover, the coloring liquids used in the present study are based on metal chloride pigments (Table 1). In agreement with the present findings, the production of colored zirconia specimens by immersion in such metal chloride pigments did not show detrimental effects on the shear bond strength to resin composite cylinders in a previous study [35]. The bond strength of zirconia restorations to resin‐based luting agents has been considered safe for clinical applications with sandblasting of AlO₂ particles after sintering and application of zirconia primers [15].

SEM images have rejected the fourth studied hypothesis. The images did not show different topographic features in specimens treated with coloring liquids, such as defects that could influence the mechanical and adhesive properties of zirconia [16, 17, 18, 19]. Grain characteristics, such as size and morphology, which could influence the materials' properties, did not differ qualitatively between control and experimental specimens. This outcome is in agreement with previous studies, which showed uniformly sized zirconia crystals after the sintering process in specimens shaded with coloring liquids [36, 37, 38].

The present study has assessed a technique to improve the substrate masking ability of zirconia restorations with the use of coloring liquids at the intaglio surface before sintering. It is important to note that such technique is particularly of interest because it does not require additional time‐consuming steps for the restoration production. However, the coloring liquid application is operator‐dependent, which means, based on the outcomes of the study, calibration of the dental ceramist is necessary to improve the substrate masking ability of zirconia restorations without negatively influencing colorimetric characteristics.

Finally, as limitations, it is worth considering that the specimen characteristics and the composition of each commercial material might interfere on the results of the study. Therefore, the outcomes of such technique might be different for tooth‐characterized restorations (instead of polished, disc‐shaped specimens of the present study) and for zirconia of other manufacturers.

5. Conclusion

Whitin the limitations of this current study, it was concluded that:

• Lightness, chroma, and hue angle of zirconia specimens showed to be influenced by the use of coloring liquids; whiteness index for dentistry was influenced mostly for thinner specimens (1.5 mm) and more translucent materials (4Y‐ and 5Y‐PSZ).

• The use of coloring liquids at zirconia intaglio surface reduces its translucency parameter and, in consequence, improves its capacity to mask discolored substrates (C3, C4, coppery and silvery metal) based on color differences in comparison with the referential (A2).

• The adhesive properties of zirconia to a resin‐based luting agent were not relevantly impaired by the use of coloring liquid, as the microshear bond strength values among groups were in the range from 20 to 25 MPa.

• The use of coloring liquids did not show to alter the topographic features and grain characteristics of zirconia in the images of scanning electron microscopy.

Funding

The present study was supported by the Brazilian Federal Agency ‘Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—CAPES’ (Finance Code 001), through grants for social demands for the Post‐graduate Program of Oral Science (UFSM).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.